Remittance behaviours among recent immigrants in Canada

Transcription

1 Remittance behaviours among recent immigrants in Canada By René Houle Research Data Center Analyst, Social and Demographic Statistics Branch, Statistics Canada and Grant Schellenberg Senior Analyst, Business and Labour Market Analysis, Statistics Canada February 6, 2008

2 I. Introduction Remittances the money immigrants send to family members in their country of origin have long been part of the immigration process. But with global networks of financial institutions and telecommunications technologies now in place, the transmission of funds worldwide now takes place at a pace and volume unimaginable by earlier generations. In this context, remittances are central in the global movement of people, information and resources and are centre stage in immigration and development research. In spite of this interest, research on the characteristics of remittances senders in Canada remains quite limited, in large part because of the absence of household survey data. More broadly, studies of remittance senders in Canada and elsewhere are often focused on immigrants from only one or two source countries and consequently do not provide a broad cross-national perspective. This study addresses these gaps by using the Longitudinal Survey of Immigrants to Canada (LSIC) to document the incidence of remitting and the amounts remitted by immigrants from a wide range of countries. Using a common set of concepts and methods, we find that the incidence of remitting by the landing cohort ranges from less than 10% to 60% across immigrants from different countries, while the average annual amounts remitted range from about $500 to almost $3,000. Turning to the factors associated with remitting, financial and family characteristics are consistently significant among immigrants from all world regions. In contrast, other factors, such as gender and education, are associated with remitting among immigrants from some regions but not others. Context Considerable work is underway both nationally and internationally to measure remittance flows (Haug 2007). Defining remittances as the sum of workers remittances, compensation of employees, and migrant transfers recorded in national balance of payment estimates, the World Bank estimates recorded remittance flows to developing countries at $167 billion US in 2005 (World Bank 2006). This is likely an underestimate as some remittances flowing through formal channels, such as transfers through post offices or exchange bureaus and remittances below a minimum threshold, are often not recorded in official estimates. Furthermore, remittances flowing through informal channels, such as those delivered by family or friends, generally go unrecorded. The World Bank estimates that unrecorded remittances could add 50% or more to the total. Remittances represent an important source of foreign revenue for developing countries. In absolute terms, India ($21.7 billion), China ($21.4 billion), and Mexico ($18.1 billion) top the list of remittance receiving countries (World Bank 2006). In proportional terms, the importance of remittances to many smaller countries is evident. For example, remittances account for about 20% to 30% of GDP in Tonga, Moldova, Lesotho, Haiti, Bosnia and Hezegovina and Jordon, and for about 10% to 19% of GDP in several others, such as Jamaica, El Salvador, the Philippines, the Dominican Republic, Lebanon and Nepal. The importance of remittances can also be documented relative to national industries. For example, the Inter-American Development Bank reports that remittances to Mexico are more 2

3 than the country s total tourism revenues, more than two-thirds of the value of petroleum exports, and about 180% of the country s agricultural exports. (Inter-American Development Bank 2004). More broadly, the World Bank reports that in 28 countries, remittances are larger than the earnings from the most important commodity export. (World Bank 2006, 88) Remittances also exceed overseas development aid and foreign direct investment in many countries. Recorded estimates of remittance flows to developing countries show a marked increase in recent years, rising by 73% between 2001 and This trend has been evident across a wide range of nations (World Bank 2006). A number of factors are likely at play, including improvements in data collection, a shift in remittances from informal to formal networks and developments within the remittance industry (World Bank 2006; Orozco 2006). Objective and Rationale The objective of this paper is to use recently released data from the Longitudinal Survey of Immigrants to Canada (LSIC) to examine the remittance activities of immigrants from different countries of origin, and to identify the socio-economic characteristics associated with such activities. To date, Canadian household data on remittances has been very limited and only a few studies have been done (for example, see Hernandez-Coss 2006; Simmons, Plaza and Piché 2005; Hamza 2006; Citizenship and Immigration Canada 2004). This stands in stark contrast to the United States where the well-developed US-Latin American remittance corridor has been the subject of considerable research over the years. Likewise, research on the remittance activities of immigrants in Australia and New Zealand dates back more than two decades. While a great deal of Canadian research continues to focus on the labour market and income characteristics of recent immigrants, 1 little attention has been paid to their expenditures, of which remittances are one component. Recent immigrants preferences and/or obligations to send money to family members abroad may have implications for other aspects of settlement, such as housing or employment decisions. And while high rates of low income among recent immigrants underscore the financial constraints often faced by new Canadians, such figures do not take into account the fact that household incomes may be used to support family members abroad. From a macroeconomic perspective, household data on remittance sending contributes to our understanding of international financial flows and plays a role in the development of concepts and measures for systems of national accounts and balance of payments (Haug 2007). Internationally, there is interest on the part of agencies such as the IMF, World Bank, and Inter- American Development Bank (IADB) in the institutional characteristics of bilateral remittance corridors. For example, Hernandez-Coss (2006) notes that efforts are underway to induce users [remittance senders] to shift from informal to formal systems in order to increase the transparency of remittance flows and enhance their contribution to development in the recipient countries. The Multilateral Investment Fund of the IADB identifies better of documentation of The two authors contributed equally to this paper. 1 Considerable emphasis has been placed on their earnings trajectories after arrival, the economic returns to their foreign credentials and experience, their ability to find employment in their area of specialization, and their incidence of low income. For a review see Þicot

4 the importance of remittances, reduced transaction costs and improved leveraging of the development impact of remittances as key objectives (Orozco 2004). Information on the entire remittance process, from the first mile when decisions are in the hands of remittances senders to the last mile when the funds are in the hands of recipients, is needed to build a complete picture of this complex phenomena. The broad cross-national perspective of remittance behaviours offered by the LSIC is a valuable contribution in this context. The paper is divided into several sections. In section two, a review of the literature is presented using an approach similar to Menjivar et al (1998). The factors potentially associated with remittance behaviours are discussed in terms of (i) financial capacity (ii) obligations to family (iii) characteristics of migration (iv) organizational involvement and (v) country of origin. In section three, the data source, methodology and variables used in the study are discussed. In section four, the results of the analysis are presented. A series of descriptive statistics is first presented, followed by the results of several multivariate models. In section five, the main conclusions and implications of our analysis are discussed. II. Literature Review and Conceptual Framework There is a large and rapidly growing literature on the impacts that remittances have in recipient countries (For example see World Bank 2007; Acosta, Fajnzylber and Lopez 2007). A review of this literature is beyond the scope of this paper, and instead we focus on studies of remittance senders. Demographic characteristics A number of studies examine the correlation between demographic characteristics and remittance behaviours. No clear pattern is found between women and men. Vanwey (2004) and Lucas and Stark (1985) find that women tend to remit more often than men while Massey and Parrado (1994) find that women remit less often. Given that these three studies pertain to women and men from Thailand, Botswana and Mexico, the inconsistent results may reflect cultural differences across countries. In terms of age, Menjivar et al (1998) find a U-shape relationship between age and the incidence of remitting; that is, they find that remitting is more prevalent among younger and older immigrants and less prevalent among middle aged immigrants. Conversely, among immigrants who remit, they find an inverted U-shape relationship between age and the average amount sent, with immigrants in their late thirties remitting the largest amounts. They suggest that this corresponds to the age-earnings profile. Financial capacity to remit A common theme in the literature on remittances is that, other things being equal, households are more likely to remit and to remit larger amounts if they have the financial capacity to do so. This is documented in terms of a positive correlation between household income and remittances, as well as full-time employment and remittances (Menjivar et al. 1998; Brown and Poirine 2005; Funkhouser, 1995). In addition, Menjivar et al. (1998) find that while home ownership is not 4

5 correlated with the likelihood of remitting, it is negatively correlated with the amount remitted. One interpretation is that home ownership is associated with higher housing costs than renting (given insurance costs, property taxes and maintenance) leaving less money to send abroad. Alternatively, individuals who purchase a home may be less likely to return to their country of origin and consequently be less likely to remit. In addition to current financial resources, remittance behaviours may also be associated with potential earnings, as measured by human capital characteristics. Here the evidence is mixed. Funkhouser (1995) finds that higher levels of educational attainment are negatively correlated with the incidence of remitting, but among migrants who do remit, those with higher levels of education send more. Massey and Basem (1992) find that human capital factors are not correlated with the decision to remit, but are correlated with the amount remitted. Conversely, Menjivar et al. (1998) find no significant correlation between human capital measures, including education and English language skills, and remittance behaviours. Conceptually, the relationship between education and remitting may run either direction. Given the positive correlation between educational attainment and labour market success, more welleducated immigrants may be more likely to remit because they have the financial capacity to do so. On the other hand, immigrants with higher levels of educational attainment may come from families that are financially better-off and hence in less need of financial support from abroad. More well-educated immigrants may also have the financial means to bring family members with them when they migrate and hence have obligations to fewer family members abroad. Obligations to family Whether or not an immigrant s children, parents or other family members are located in the country of origin or are present in Canada is a key determinant of remittances. A number of studies report that remittances are more likely to be sent when close family members still reside in the country of origin (for example, Vanwey 2004; Funkhouser 1995; Menjivar et al. 1998; Stanwix and Connell 1995; Vete 1995;). Conversely, remittance activity is negatively associated with the presence of family members in the host country (Menjivar et al. 1998). In addition to the location of family members abroad, remittance behaviours may also be influenced by the financial circumstances of those members. Acosta et al. (2007) document the income characteristics of remittance-receiving households in 11 Latin American countries. They find that in many cases, remittance-receiving households are concentrated at the bottom of the (non-remittance) income distribution. In Mexico, for example, 61% of remittance receiving households are in the bottom income quintile, while in Paraguay the share is 42%. Other Latin American countries in which remittance recipients tend to be poor include El Salvador, Guatemala, Ecuador and Paraguay. In other countries, remittance recipients tend to be more evenly distributed across income quintiles (e.g. Honduras, the Dominican Republic), while in others, notably Peru and Nicaragua, remittance recipients tend to be at the top of the income distribution (Acosta et al. 2007). Overall, these data point to many instances where remittances are received by families in financial need, but they also underscore cross-national variations in this regard. 5

6 A number of macroeconomic studies find a positive relationship between remittance flows and economic opportunities or financial hardships in countries of origin (Bouhga-Hagbe 2006; Niimi and Ozden 2006). For example, in their 85-country study, Niimi and Ozden report that small domestic economies offer relatively limited opportunities for economic activities, and migrants have to remit more to support their families at home and that migrants from poorer countries remit a greater amount of money to their families at home as would be expected. 2 In this context, it is often argued that remittance flows are countercyclical, with migrants abroad sending more money to family members in their home country when economic hardships occur there. However, a comparison of remittance flows into 12 developing countries over their business cycles between 1976 and 2003 shows that countercyclicality of receipts is not commonly observed. (Sayan 2006). Characteristics of migration The circumstances and characteristics of migration may influence remittance behaviours in a number of ways, including the motivations for migrating, intentions to return to the country of origin, the duration of time since immigration, and the number of emigrants leaving the source household. The reasons and circumstances for leaving the country of origin may be correlated with remittance behaviours. Individuals displaced from their country of origin by war or persecution may leave on short notice and not have the opportunity or means to bring other family members with them. Hence, they may have responsibilities to support those left behind. Likewise, people who migrate in search of economic opportunities may be financially responsible for dependent children or parents in the country of origin (Menjivar et al. 1998). Although Menjivar et al. (1998) hypothesize that these factors would be positively correlated with remitting, they find no such correlation among Salvadoran and Filipino immigrants. Whether migrants intend to return to their country of origin is another factor associated with remittance behaviours. Temporary workers are generally believed to remit a larger share of their income than permanent migrants (World Bank 2006, 92-93; Vete 1995; Diaz-Briquets and Perez- Lopez 1997). This may reflect a concerted economic strategy on the part of families who opt to send members abroad for a limited period of time to bolster the family s financial resources. Migrants who intend to return home may also remit in order to invest in their own financial future or to improve their social status (Ali 2007) or marriage prospects upon their return (Xiang 2001). Brown and Poirine (2005) find that the likelihood of remitting and the amounts remitted to children and parents are both significantly associated with intentions to return home. Similarly, Menjivar et al. (1998) find that plans for a permanent stay in the host country are negatively correlated with the decision to remit, but are not significantly correlated with the amount remitted. Duration of time in the host country is another migration characteristic often associated with remittances. The Inter-American Development Bank reports that among Latino migrants in the United States, the incidence of remitting is highest among those who have arrived most recently (Inter-American Development Bank, 2004). Immigrants who have been in the host country for 2 Also see Bouhga-Hagbe (2006). 6

7 longer periods may be less likely to remit because ties to the country of origin have become attenuated or because other family members have had enough time to join them in the host country. Menjivar et al. (1998) find that duration of time in the US is negatively correlated with the decision to remit, but is not correlated with the amount remitted. Funkhouser (1995) finds mixed results among El Salvadoran and Nicaraguan immigrants in the US, depending, in part, on their relationship with the recipient household. The migration behaviours of households in the country of origin are a final consideration in remittances. Specifically, Funkhouser (1995) finds that when there are more adult emigrants from the same household [in the country of origin], the first reported emigrant is less likely to remit and remits less, all else equal. The household from which more adults have emigrated, however, tends to receive more money in total from abroad. (p. 141). Organizational involvement In addition to the familial ties that immigrants retain with their country of origin, other ties may also exist. For example, a number of researchers document the organizational linkages that some migrant groups maintain with their countries and communities of origin (Orozco 2002; Levitt 1997). Hometown associations are one example, comprised of migrants who coordinate their efforts to support objectives in their country of origin, such as fundraising for charities and infrastructural development (e.g. parks, health equipment and libraries). Orozco suggests that hometown associations are formed among remittance senders to coordinate their support not only of relatives, but also of their towns. (Orozco, 2002, 48). Similarly, Simmons et al. (2005) note the role that church-to-church support programs play in the remittance activities of Jamaicans in Toronto and Haitians in Montréal. Region / Country of origin In addition to the factors considered above, remittance behaviours may also vary in terms of immigrants countries of origin. There are large differences in the institutional characteristics of remittance corridors and the ease and transparency with which funds may be sent. The World Bank estimates that over 80% of remittances sent to the Dominican Republic, Guatemala and El Salvador are sent through formal channels, such as banks and money transfer companies like Western Union, while over half of remittances sent to Bangladesh and Uganda flow through informal channels, such as unregulated firms and family and friends (World Bank 2006). In perhaps the only Canadian study on this issue, Hernandez-Coss (2005) notes that the Canada- Vietnam remittance corridor is small in absolute terms and still at a nascent stage of shifting from informal to formal systems. (p. 3) Costs vary as well. Orozco (2006) reports that within the well-developed remittance corridors between the US and Latin America, the costs of sending US$200 varies from 3.9% to Ecuador, 6.0% to Mexico, 8.2% to Jamaica and 12.0% to Cuba. The number of firms operating in the money transfer industries of these countries varies as well (Orozco 2006). Macroeconomic studies have examined the correlation between remittance flows and other national characteristics, such as interest rate and exchange rate differentials between remittance 7

8 sending and receiving countries, and the level of financial sector development in receiving countries (For example, El-Sakka and McNabb, 1999; Freund and Spatafora 2005; Niimi and Ozden 2006). Debate over the impact of these characteristics on remittance flows is ongoing. Cultural differences between countries of origin may also influence remittance behaviours. For example, Vanwey (2004) finds that Thai women are more likely than Thai men to remit funds and argues that this is consistent with religious and cultural norms regarding gender roles. The motivations for sending money to family members back home is a final theme in the remittance literature (Brown and Poirine 2005; Vanwey 2004; Lucas and Stark 1985; Stark and Lucas 1988); Lillard and Willis 1997). Theoretical approaches to this issue can be broadly classified into those based on altruism and those based on self-interest. Theories of altruistic behaviour posit that migrants send remittances home to care for family members and do so even to the detriment of their own standard of living. Theories of self-interest posit that remittances are sent home in exchange for other benefits or are a repayment (or prepayment) of debt between family members. For example, remittances may be prepayment made by adult children to their parents in anticipation of future returns through inheritance or a repayment to family members for previous investments in education or the costs of migration. Several authors have charted a middle course between these positions. For example, Brown and Poirine (2005) advance the notion of weak altruism whereby intra-family exchanges are based on a benevolent disposition of parents towards children and the loyalty of children towards parents. Similarly, Lucas and Stark (1985) advance a model of tempered altruism or enlightened self-interest where exchanges are based on implicit understandings of mutual benefit. LSIC respondents were not asked about their motivations for sending abroad to family or friends and hence this issue is not addressed in this paper. 3 III. Data source and methodology Data for this study was drawn from the Longitudinal Survey of Immigrants to Canada (LSIC). The target population of the survey, which was conducted jointly by Statistics Canada and Citizenship and Immigration Canada (CIC), includes all immigrants who (1) arrived in Canada between October 1, 2000 and September 30, 2001; (2) were age 15 or older at the time of landing; and (3) landed from abroad and applied through a Canadian Mission Abroad. The sampling frame for the LSIC was an administrative database of all landed immigrants to Canada maintained by Citizenship and Immigration Canada. The LSIC sample was created using a two-stage stratified sampling method. The first stage involved the selection of Immigrating Units (IU) using a probability proportional to size method, and the second stage involved the random selection of one member within each IU. Only the selected member was followed throughout the survey. Individuals aged 15 or older in each IU were eligible to be selected as the respondent. 3 Studies of motivations for remitting often use information on the characteristics of remittance sends and receivers. The latter is not available on the LSIC further limiting the scope for examining this issue. 8

9 Three LSIC questionnaires were fielded during the course of the survey. Approximately 12,000 immigrants were interviewed between April 2001 and March 2002, about six months after their arrival in Canada; approximately 9,300 of the same immigrants were relocated and interviewed in 2003, about two years after their arrival; and about 7,700 of the same immigrants were relocated and interviewed a third time, about four years after their arrival. The approximately 7,700 LSIC respondents who were relocated over the three waves are nationally representative of approximately 157,600 new immigrants, of whom 104,400 are economic immigrants, 42,600 are family class immigrants and 9,700 are refugees. 4 During the second and third LSIC interviews, respondents were asked: Since your last interview, have you sent money outside Canada to relatives or friends? Respondents who said yes were subsequently asked: How much money have you sent outside Canada to relatives or friends? During the first LSIC interview (6 months after arrival) respondents were only asked if, since arriving in Canada, they had sent money outside Canada to relatives or friends -- they were not asked about the amount sent. In the research literature, remittances are predominantly discussed in terms of money sent to family members rather than friends abroad. Consequently, we limit our analysis to LSIC respondents who have family members living outside Canada. This results in the exclusion of 256 of the 7,716 respondents, representing 4,909 of the 157,615 immigrants in the landing cohort. Their exclusion has virtually no effect on model estimates. In addition, respondents who remitted less than $100 or more than $25,000 are excluded to reduce the effects of outliers on model estimates. This results in the exclusion of another 26 cases, resulting in a final sample of 7,434 respondents. Because remittance questions were asked at the individual- rather than family-level (that is, have you sent money? ), the LSIC may yield underestimates of remittance activities of immigrant families. This is because it is possible that while an LSIC respondent did not send money abroad, someone else in their family did. This may be most likely when the respondent is not a primary income earner in the family (e.g. teens living with their parents or seniors living with their adult children). A variable identifying principal applicants and spouses and dependents in the migrating unit was initially included in our analysis, but was dropped but it did not yield significant results. Instead, a variable identifying whether or not the respondent is 4 Individuals who applied and landed from within Canada were excluded from the survey since they may have been in the country for a considerable length of time before landing and may therefore demonstrate different integration characteristics than those arriving more recently. Refugees claiming asylum from within Canada were also excluded from the survey. Interviews were conducted in one of 15 languages covering approximately 93% of the new immigrant population in Canada and were conducted face-to-face, or by telephone when a face-to-face interview was not possible. For more information on the LSIC see the microdata users guide available at 9

10 the person most knowledgeable about the family s financial situation is included. About threequarters of our respondents (73.5%) identify themselves as the person most knowledgeable. Three other points regarding our sample are warranted. First, the question of whether remittance activities diminish over time since arrival in the host country can only be addressed within the four year window of the survey. Data sources are not available to determine the extent to which immigrants in Canada continue to remit five, ten or fifteen years after arrival. Second, temporary residents in Canada are not included in the LSIC. The objective of the survey is to track the experiences of permanent residents over their first four years here and short-term, temporary residents fall outside the scope of the survey. Hence, we are not able to compare the remittance behaviours of temporary and permanent residents. In 2005, there were approximately 160,000 foreign nationals residing in Canada on temporary work permits (CIC 2007). Finally, it is important to note that the characteristics and experiences of the LSIC immigrant cohort may or may not be the same as those of cohorts that arrived in Canada at earlier or later points in time. For example, the downturn of the high-technology sector may have had particular consequences for immigrants arriving in , including their financial capacity to remit. While it would be informative to estimate inter-cohort trends or differences in remittance activities, our data do allow us to do so. While recognizing these limitations, the LSIC offers other advantages. It includes a large, representative sample of new immigrants from many source countries, making it possible to draw cross-national comparisons of remittance behaviours using consistent definitions and methodologies. This provides a breadth of perspective not available from studies focused on immigrants from one or two countries of origin. Furthermore, the longitudinal design of the LSIC allows us to build in a time-lag between our independent and dependent variables thereby minimizing problems of endogeneity. More specifically, two observations are available for each respondent -- remittances made (or not made) two years after arrival and four years after arrival. For the observation two years after arrival, independent variables are measured using the sixmonth data file, and for the observation four years after arrival, independent variables are measured using the two-year file. In short, the independent variables are measured at the start of the reference period and thus precede the decision to remit. 5 Independent variables Independent variables within each of the five conceptual areas outline above are included in the model. 6 Financial capacity is measured in terms of family income, employment status of the respondent and the value of savings abroad. A flag identifying whether or not the respondent is the person most knowledgeable about the family s income is also included, as noted above. 7 5 One exception is income, which is estimated at the mid-point of each of the reference periods using a linear interpolation. 6 Our initial multivariate models included more than 20 independent variables, many of which did not yield significant statistical results or theoretical insights. They were subsequently excluded from our analysis for the sake of efficiency and parsimony. The variables which were excluded are documented in the footnotes to this section for readers who may be interested. 7 In earlier versions of the analysis, housing expenditures as a percent of family income was also included. A significant negative correlation was found between such expenditures and both the likelihood of remitting and the amount remitted. However, the housing choices made by an immigrant may be influenced by their decision to remit 10

11 Human capital is measured by level of education upon landing in Canada. Self-assessed language ability in English or French was included in earlier versions of the analysis, but was subsequently excluded because it did not yield significant results. A place of residence variable identifying immigrants residing in Montréal, Toronto, Vancouver, Calgary/Edmonton and others areas is included. Detailed information on the location of all the members of each immigrant s family is not available on the LSIC. The vast majority of LSIC respondents (98%) say they have family members living abroad, but the relationships with these members are not specified. Two familyrelated variables are included in our models: the number of children residing in the respondent s household and whether or not the respondent is sponsoring or intending to sponsor (i) a spouse or child(ren) or (ii) parents or grandparents to come to Canada. 8 In terms of migration characteristics, the category of admission to Canada is included in our models. Individuals are admitted to Canada as permanent residents through three main admission categories -- economic immigrants, family class immigrants and refugees. Principal applicants in the economic category are selected for their skills and ability to contribute to Canada s economy and are assessed on the basis of educational attainment, language abilities and other factors. Because the earnings gains of economic immigrants far outpace those of family class immigrants and refugees in the years after arrival (Chui and Tran 2003), they may be more likely to remit and to remit larger amounts given their greater financial capacity to do so. 9 In terms of organizational membership, LSIC respondents are asked if they had become a member or taken part in the activities of various groups and organizations since their last interview. Two dummy variables are included identifying respondents who had or had not participated in or become a member of a (i) church or other religious group or (ii) an ethnic or immigrant association. 10 Region of origin is included in our analysis in two ways. First, regression models based on a pooled sample of all respondents are presented, with nine world regions identified by a series of dummy variables. And second, separate regression models are run on respondents from each of these nine regions. Finally, Gross Domestic Product (GDP) per capita in the respondent s country of origin is included. or by their expectation of remitting in the future. Hence, housing expenditures may be a consequence of remittance decisions rather than a determinant of these decisions. Because of this potential endogeneity, the housing expenditure variable was excluded from our final models. 8 Respondents sponsoring or intending to sponsor a spouse or child as well as a parent or grandparent were coded in the sponsoring spouse/child category. There were fewer than 50 respondents in this situation. Marital status was included in earlier versions of the analysis, but did not yield significant results and was subsequently excluded. 9 In earlier versions of the multivariate models, a variable was included identifying the respondent s single-most important reason for coming to Canada; specifically economic reasons, family-related reasons, reasons of war or political freedoms, and other reasons. These variables did not yield significant results and were excluded. 10 A third variable identifying participation in any other type of group or organization was also initially included, but did not yield significant results. 11

12 Our multivariate analysis includes a logistic regression on the likelihood of remitting and an ordinary-least squares regression on the natural logarithm of the amount remitted. Coefficients from the logistic regressions have been converted into predicted probabilities for ease of interpretation. 11 Coefficients from the natural logarithm of the amount remitted approximate percentage differences and are discussed in these terms for ease of presentation. All models are calculated using bootstrap weights to correct variance estimates for survey design (a technique called design-based variance estimation). 12 IV. Results Descriptive results A significant minority of immigrants from the landing cohort remitted funds to family or friends abroad during their first four years in Canada. During the period 6 to 24 months after landing, 23% of immigrants remitted and during the period 25 to 48 months after landing 29% did so (Table 1). Readers are reminded that these two references periods are not of equal length (i.e. 18 and 24 months respectively). Combined information from the 1 st and 2 nd waves of the LSIC indicates that 28% of immigrants remitted during the first 24 months after landing almost the same proportion that did so during the subsequent 24-month period. Finally, 41% of LSIC respondents remitted funds at least once during their first 48 months in Canada. Among immigrants who remitted, the average amount sent during the first reference period (6 to 24 months after landing) was $2,500, while the average amount sent by remitters during the second reference period (25 to 48 months after landing) was $2, These averages correspond to reference periods of different duration. Assuming the total amount sent was evenly distributed over the reference period, annual remittances during the third and fourth year in Canada were $1,450. This is comparable to the estimates reported by Simmons et al. (2005), who find that Haitian and Jamaican immigrants who remit send approximately $1,000 to $1,400 per year. The extent to which remittance behaviours vary between immigrants from different regions of birth is clearly evident in Table 1. Over half of LSIC respondents from Southeast Asia and the 11 Predicted probabilities for each independent variable were estimated by setting the other independent variables to their mean values. 12 Some researchers have used the Heckman selection model (1976) to take into account the possibility that the sample of immigrants who remit may be a selective sample of those who could have remitted (Funkhouser 1995; Brown and Piorine 2005). Several Heckman models were run using different specifications to address this issue but evidence of selectivity was not found. Our results are consistent with several studies that also report that selection effects are modest or not statistically significant (Menjivar et al. 1998; Funkhouser 1995). 13 All dollar figures have been rounded to the nearest $100. Remittance amounts reported 2 years and 4 years after arrival have not been adjusted for inflation. Questions about remitting and remittance amounts were included in the income section of the LSIC questionnaire. This section includes numerous questions about the income of the respondent and respondent s family all of which refer to the 12 month period preceding the interview. At the end of the section, respondents were asked if they had remitted since their last interview, and if so, how much they had remitted. Here, the reference period shifts from the 12 months preceding the survey to the 18 or 24 month period preceding the survey (the duration varies between Waves 2 and 3). Given the sudden shift in the reference periods, we cannot be certain if respondents who reported remittance amounts had a 12-month or 18/24-month reference period in mind. 12

13 Caribbean and Guyana sent remittances home 25 to 48 months after landing, while the was the case for about 40% of those from sub-saharan Africa and Eastern Europe. About one-quarter of the respondents from South Asian and Central and South America sent remittances home during this period, while about one-fifth of those from East Asia and from West Asia, the Middle East and North Africa did so. Among immigrants who remitted 25 to 48 months after landing, the average amount sent was $2,900, but again, inter-regional differences are evident. Immigrants from East Asia who remitted sent, on average, $3,900 over the 24 month reference period, while immigrants from the Caribbean and Guyana sent, on average, less than half that amount ($1,600). Among LSIC respondents, the incidence of remitting is highest among those from countries with lower GDP per capita. Over the period 25 to 48 months after landing, 35% to 37% of immigrants from countries with GDP per capita under $4,000 remitted, compared with only 11% of immigrants from countries with GDP per capita of $15,000 or more. One interpretation is that the family members of immigrants from poorer countries are in greater need of financial support than the family members of immigrants from more affluent countries, and hence the former are more likely to remit. That being said, the relationship between GDP per capita and the incidence of remitting is fairly flat between these extremes, ranging from about 25% to 30%. Conditional on remitting, a consistent relationship between GDP per capita and average amounts remitted is not evident in the descriptive statistics shown in Table 1. 13

14 Table 1. Percent of LSIC respondents who remitted and average amount remitted* 6 to 24 mths after arrival 25 to 48 mths after arrival Average over both periods % who remitted Average amt remitted* % who remitted Average amt remitted* % who remitted Average amt remitted* Total 23% $2,500 29% $2,900 26% $2,700 Region of birth Southeast Asia 52 $2, $2, $2,200 Caribbean & Guyana 47 $1, $1, $1,500 Sub-Saharan Africa 37 $2, $2, $2,500 East Europe 32 $1, $2, $1,900 South Asia 23 $3, $3, $3,600 Central & South America 23 $2, $2, $2,000 East Asia 13 $2, $3, $3,500 West Asia, Mid-East & N. Africa 13 $2, $2, $2,300 N. America, W. Europe, Oceania 11 $3, $3, $3,400 GDP/capita in country of birth Less than $2, $1, $2, $2,100 $2,000 to $3, $2, $3, $2,800 $4,000 to $5, $2, $3, $2,900 $6,000 to $7, $1, $2, $2,000 $8,000 to $14, $2, $1, $2,100 $15,000 and over 8 $3, $3,900 9 $3,500 * Average amount reported by those who remitted. Estimates rounded to nearest $100. Remittance values reported two years and four years after arrival have not been adjusted for inflation.

15 Turning to comparisons across countries of birth, variability in remittance behaviours is particularly striking. As shown in Chart 1, 60% of immigrants from the Philippines and Haiti remitted two to four years after landing, 14 while about 40% to 50% of immigrants from Jamaica, Nigeria, Romania, Guyana and the Ukraine did so. Quite clearly, remittances are sent by many new immigrants from a diverse set of world regions. France, the United Kingdom and South Korea three industrialized countries are at the bottom of the distribution shown in Chart 1. Turning to the average amounts of money sent abroad, remitters from 11 of the 24 countries shown in Chart 2 sent between $1,700 and $2,200, and remitters from another 7 countries sent between $2,700 and $3,700. While less than 20% of immigrants from the United States sent remittances home, the average amount sent by these individuals was quite high relative to immigrants from other countries, at just under $6,000. Readers should note, however, that the confidence intervals around many of the estimates are quite large. 14 The estimates in Charts 1 are computed by taking the average of the incidences of remitting at LSIC Wave 2 (i.e. 24 months after landing) and at LSIC Wave 3 (i.e. 48 months after landing). This approach reduces standard errors around the estimates (which are still large in many cases) and simplifies the presentation of the data. The same approach is used for Chart 2.

16 70% Chart 1. Percent of LSIC respondents who remitted two to four years after landing*, by selected countries of birth 60% 50% 40% 30% 20% 10% 0% Philllipines Haiti Jamaica Nigeria Romania Guyana Ukraine Russia Colombia Sri Lanka Bosnia Pakistan India Lebanon China Morocco USA Mexico Afghanistan Iraq Iran France UK Average of the incidences of remitting at tw o years and four years after landing. Vertical bars show 95% confidence interval around the estimate. S. Korea $9,000 Chart 2. Average amount remitted two to four years after landing*, by selected countries of birth $8,000 $7,000 $6,000 $5,000 $4,000 $3,000 $2,000 $1,000 $0 Philippines Haiti Jamaica Nigeria Romania Guyana Ukraine Russia Colombia Sri Lanka Bosnia Pakistan India Lebanon China Morocco USA Mexico Afghanistan Iraq Iran France Average of the average amounts remitted (conditional on remitting) at tw o years and four years after landing. Vertical bars show 95% confidence interval around the estimate. UK S. Korea 16

17 In terms of the admission categories through which immigrants are admitted to Canada, about 30% of immigrants in all three categories remitted 25 to 48 months after landing (Table 2). However, among the individuals who did remit, economic immigrants sent somewhat larger amounts than refugees (at $3,000 and $1,900 respectively). Measures of central tendencies, such as averages, tell us little about the range of remittance values. As shown Table 2, 26% of immigrants who remitted during the period 25 to 48 months after landing sent less than $500. This was the case for 21% of economic immigrants compared with 45% of refugees. About one-half of immigrants in all categories sent between $500 and $2,500. And at the high end of the distribution, 12% of the economic immigrants who remitted sent $5,000 or more compared with 5% of refugees. Table 2. Characteristics of remittances 25 to 48 months after landing, by immigrant category Immigrant category Total Economic Family-class Refugees Percent who remitted 29% 29% 29% 31% Average amount remitted (CAN$) $2,900 $3,000 $2,700 $1,900 Of those remitting, % who sent Less than $ $500 to $ $1,000 to $2, $2,500 to $4, $5,000 or more Total 100% 100% 100% 100% An important issue is the extent to which remittance activities impose financial hardships on newly arrived immigrants. Several studies have documented the relatively high and rising rates of low-income among recent immigrants (Heisz and McLeod 2004; Picot, Hou and Coulombe 2007). While measures of low-income take into account the number of family members coresiding together, they do not take into account the sharing of income with family members residing outside of the household, either in Canada or abroad. This applies to all families regardless of immigration status. However, given the relatively high rates of low income among recent immigrants and the fact that almost one-third of them send money abroad, their financial resources may be stretched further than income figures alone would suggest. A cautious approach is warranted when addressing this issue. As noted above, because remittance behaviour is measured using individuals rather than families or households as the unit of analysis, estimates of amounts sent abroad are likely to be conservative. Furthermore, remittances as a share of income can be computed using total personal income or total economic family income as the denominator. Total personal income yields a higher percentage, but does not take into account any sharing of financial resources among family members. Total family income yields a lower percentage, but mixes units of analysis (personal remittances and family income). Results from both approaches are presented in Table 3 and represent conservative estimates of the lower and 17

18 upper bounds of remittances as a share of income. During the second year in Canada, remittances accounted for 7.5% of the total personal income of remitters and for 3.4% of their total family income, on an average annualized basis. Two years later, remittances accounted for 5.9% and 2.9% of the total personal and total family incomes of remitters. 15 Table 3. Characteristics of remittances on an annualized basis, by immigrant category Total Economic Family Refugees 2 nd year after arrival Of those who remitted Average total personal income $22,200 $27,200 $14,100 $12,500 Average amount remitted as % of average total personal income 7.5% 6.7% 10.4% 8.1% Average total family income $48,700 $51,00 $47,400 $28,300 Average amount remitted as % of average total family income 3.4% 3.5% 3.1% 3.6% 4 th year after arrival Of those who remitted Average total personal income $28,200 $33,600 $17,500 $16,400 Average amount remitted as % of average total personal income 5.9% 5.4% 8.4% 6.1% Average total family income $57,200 $61,100 $52,600 $36,100 Average amount remitted as % of average total family income 2.9% 3.0% 2.8% 2.8% Considering all immigrants in our sample regardless of whether or not they remitted, remittances accounted for 3.7% and 3.4% of total aggregate personal income and for 1.6% and 1.3% of total aggregate family income two and four years after arrival. From this perspective, remittances account for a fairly small share of the total aggregate income of newly arrived immigrants. Nonetheless, remittances may still represent a considerable expenditure for some families. Take refugees for example. The average family incomes of refugees who remitted during their fourth year in Canada was $36,100. As a point of comparison, the 2004 before-tax low-income cut-off (LICO) was just over $31,000 for a three-person family and just under $38,000 for a four-person family residing in a large urban centre. The LICO is an income threshold below which a family will likely devote a larger share of its income on the necessities of food, shelter and clothing than the average family (Statistics Canada 2006). For those refugees who remitted, an average of almost $1000 was expended from a fairly modest family income. 15 For immigrants who remitted we also computed average total family income after expenditures on housing (including rent or mortgage, taxes and utilities), and used this to estimate remittances as a share of family income after expenditures on housing. For immigrants in all three admissions categories, remittances accounted for about 4.0% to 4.9% of family income after expenditures on housing. 18

19 Multivariate results Pooled model Descriptive statistics certainly testify to the magnitude of cross-national differences in remittance behaviours. These differences partly reflect the different characteristics and experiences of individuals from different countries of origin, and it is to these that we now turn. We begin by presenting results from a logistic regression and ordinary least squares regression based on our pooled sample. The compositional characteristics of immigrants in the pooled sample are shown in Table 5. Demographic characteristics Considering demographic characteristics, there is a modest correlation between sex and remittance behaviours (Table 4), with predicted probabilities of remitting (after taking into account other observed characteristics) of 26% for men and 23% for women. Among those who remit, the amount sent by women is approximately 12% less than the amount sent by men. Age is also important, as the predicted probability of remitting is highest among immigrants aged 25 to 44 (at about 30%) and lower among those in younger and older age groups (at less than 20%). Conditional on remitting, individuals age 25 to 34 send larger amounts than those under age 25 or age 65 or older. Financial capacity to remit Consistent with the literature, there are strong correlations between remittance behaviours and financial capacity. For example, the predicted probability of remitting rises monotonically across family income categories, from 10% among immigrants in families with incomes under $10,000 to 36% among those in families with incomes of $70,000 or more. Conditional upon remitting, the amounts sent abroad also increase monotonically across income categories. The amount sent abroad by remitters in families with incomes of $70,000 or more is approximately 45% higher than the amount sent by remitters in families with incomes of $25,000 to $44,999. Considering savings abroad, immigrants who have $5000 or more in savings outside of Canada are significantly less likely to remit (at about 20%) than immigrants who have no savings abroad (at 26%). One interpretation is that immigrants with savings abroad come from more affluent families than those who have no savings abroad, and hence the former are less likely to remit. Among immigrants who remit, the amount sent abroad is not correlated with savings. Consistent with other studies, immigrants who are employed on a full-time basis are significantly more likely to remit than those who are employed part-time or are not in the labour force (predicted probabilities of 29%, 25% and 21% respectively). However, employment status is not correlated with the amount sent. The probability of remitting is not significantly associated with the level of education that immigrants had when they arrived in Canada. However, conditional upon remitting, the amounts sent abroad by immigrants with high school or less are 20% to 25% less than the amounts sent by those with a university degree. 19